We study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another. In particular, we apply classes assigned to YouTube videos to tweets linking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The experiments we have run show that a tweet classifier trained via these automatically labelled data substantially outperforms an analogous classifier trained with a limited amount of manually labelled data.
Distant supervision for tweet classification using YouTube labels
Sebastiani F
2015
Abstract
We study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another. In particular, we apply classes assigned to YouTube videos to tweets linking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The experiments we have run show that a tweet classifier trained via these automatically labelled data substantially outperforms an analogous classifier trained with a limited amount of manually labelled data.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
prod_331191-doc_102179.pdf
solo utenti autorizzati
Descrizione: Distant supervision for tweet classification using YouTube labels
Tipologia:
Versione Editoriale (PDF)
Dimensione
465.11 kB
Formato
Adobe PDF
|
465.11 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
prod_331191-doc_159215.pdf
accesso aperto
Descrizione: Distant supervision for tweet classification using YouTube labels
Tipologia:
Versione Editoriale (PDF)
Dimensione
232.98 kB
Formato
Adobe PDF
|
232.98 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.